Hundreds of medically useful compounds are discovered each year, but clinical use of these drugs is possible only if a drug delivery vehicle is developed to transport them to their therapeutic target in the human body. This problem is particularly critical for drugs requiring intravenous injection in order to reach their therapeutic target or dosage but which are water insoluble or poorly water soluble. For such hydrophobic compounds, direct injection may be impossible or highly dangerous, and can result in hemolysis, phlebitis, hypersensitivity, organ failure and/or death. Such compounds are termed by pharmacists “lipophilic”, “hydrophobic”, or in their most difficult form, “amphiphobic”.
A few examples of therapeutic substances in these categories are ibuprofen, diazepam, griseofulvin, cyclosporin, cortisone, proleukin, etoposide and paclitaxel. Kagkadis, K A et al. (1996) PDA J Pharm Sci Tech 50(5):317–323; Dardel, O. 1976. Anaesth Scand 20:221–24. Sweetana, S and M J U Akers. (1996) PDA J Pharm Sci Tech 50(5):330–342.
Administration of chemotherapeutic or anti-cancer agents is particularly problematic. Low solubility anti-cancer agents are difficult to solubilize and supply at therapeutically useful levels. On the other hand, water-soluble anti-cancer agents are generally taken up by both cancer and non-cancer cells thereby exhibiting non-specificity.
Efforts to improve water-solubility and comfort of administration of such agents have not solved, and may have worsened, the two fundamental problems of cancer chemotherapy: 1) non-specific toxicity and 2) rapid clearance form the bloodstream by non-specific mechanisms. In the case of cytotoxins, which form the majority of currently available chemotherapies, these two problems are clearly related. Whenever the therapeutic is taken up by noncancerous cells, a diminished amount of the drug remains available to treat the cancer, and more importantly, the normal cell ingesting the drug is killed.
To be effective in treating cancer, the chemotherapeutic must be present throughout the affected tissue(s) at high concentration for a sustained period of time so that it may be taken up by the cancer cells, but not at so high a concentration that normal cells are injured beyond repair. Obviously, water soluble molecules can be administered in this way, but only by slow, continuous infusion and monitoring, aspects which entail great difficulty, expense and inconvenience.
A more effective method of administering a cancer therapeutic, particularly a cytotoxin, is in the form of a dispersion of oil in which the drug is dissolved. These oily particles are made electrically neutral and coated in such a way that they do not interact with plasma proteins and are not trapped by the reticuloendothelial system (RES), instead remaining intact in the tissue or blood for hours, days or even weeks. In most cases, it is desirable if the particles also distribute themselves into the surrounding lymph nodes which are injected at the site of a cancer. Nakamoto, Y et al. (1975) Chem Pharm Bull 23(10):2232–2238. Takahashi, T et al. (1977) Tohoku J Exp Med 123:235–246. In many cases direct injection into blood is the route of choice for administration. Even more preferable, following intravenous injection, the blood-borne particles may be preferentially captured and ingested by the cancer cells themselves. An added advantage of a particulate emulsion for the delivery of a chemotherapeutic is the widespread property of surfactants used in emulsions to overcome multidrug resistance.
For drugs that cannot be formulated as an aqueous solution, emulsions have typically been most cost-effective and gentle to administer, although there have been serious problems with making them sterile and endotoxin free so that they may be administered by intravenous injection. The oils typically used for pharmaceutical emulsions include saponifiable oils from the family of triglycerides, for example, soybean oil, sesame seed oil, cottonseed oil, safflower oil and the like. Hansrani, P K et al., (1983) J Parenter Sci Technol 37:145–150. One or more surfactants are used to stabilize the emulsion, and excipients are added to render the emulsion more biocompatible, stable and less toxic. Lecithin from egg yolks or soybeans is a commonly used surfactant. Sterile manufacturing can be accomplished by absolute sterilization of all the components before manufacture, followed by absolutely aseptic technique in all stages of manufacture. However, improved ease of manufacture and assurance of sterility is obtained by terminal sterilization following sanitary manufacture, either by heat or by filtration. Unfortunately, not all emulsions are suitable for heat or filtration treatments.
Stability has been shown to be influenced by the size and homogeneity of the emulsion. The preferred emulsion consists of a suspension of sub-micron particles, with a mean size of no greater than 200 nanometers. A stable dispersion in this size range is not easily achieved, but has the benefit that it is expected to circulate longer in the bloodstream. Further, less of the stable dispersion is phagocytized non-specifically by the reticuloendothelial system. As a result the drug is more likely to reach its therapeutic target. Thus, a preferred drug emulsion will be designed to be actively taken up by the target cell or organ, and is targeted away from the RES.
The use of vitamin E in emulsions is known. In addition to the hundreds of examples where vitamin E in small quantities [for example, less than 1%, RT Lyons. Pharm Res 13(9): S-226, (1996) “Formulation development of an injectable oil-in-water emulsion containing the lipophilic antioxidants K-tocopherol and P-carotene”] is used as an anti-oxidant in emulsions, the first primitive, injectable vitamin E emulsions per se were made by Hidiroglou for dietary supplementation in sheep and for research on the pharmacokinetics of vitamin E and its derivatives. Hidiroglou M and Karpinski K. (1988) Brit J Nutrit 59:509–518.
For mice, an injectable form of vitamin E was prepared by Kato and coworkers. Kato Y., et al. (1993) Chem Pharm Bull 41(3):599–604. Micellar solutions were formulated with Tween 80, Brij 58 and HCO-60. Isopropanol was used as a co-solvent, and was then removed by vacuum evaporation; the residual oil glass was then taken up in water with vortexing as a micellar suspension. An emulsion was also prepared by dissolving vitamin E with soy phosphatidycholine (lecithin) and soybean oil. Water was added and the emulsion prepared with sonication.
In 1983, E-Ferol, a vitamin E emulsion was introduced for vitamin E supplementation and therapy in neonates. Alade S L et al. (1986) Pediatrics 77(4):593–597. Within a few months over 30 babies had died as a result of receiving the product, and the product was promptly withdrawn by FDA order. The surfactant mixture used in E-Ferol to emulsify 25 mg/mL vitamin E consisted of 9% Tween 80 and 1% Tween 20. These surfactants seem ultimately to have been responsible for the unfortunate deaths. This experience illustrates the need for improved formulations and the importance of selecting suitable biocompatible surfactants and carefully monitoring their levels in parenteral emulsions.
An alternative means of solubilizing low solubility compounds is direct solubilization in a non-aqueous milieu, for example alcohol (such as ethanol) dimethylsulfoxide or triacetin. An example in PCT application WO 95/11039 describes the use of vitamin E and the vitamin E derivative TPGS in combination with ethanol and the immuno-suppressant molecule cyclosporin. Alcohol-containing solutions can be administered with care, but are typically given by intravenous drip to avoid the pain, vascular irritation and toxicity associated with bolus injection of these solutions.
Problems with pharmaceutical formulations in non-aqueous solvents and solubilizers such as alcohol (ethanol, isopropanol, benzyl alcohol, and the like) relate to the ability of these solvents to extract toxic substances, for example plasticizers, from their containers. The current commercial formulation for the anti-cancer drug paclitaxel, for example, consists of a mixture of hydroxylated castor oil and ethanol, and rapidly extracts plasticizers such as di-(2-ethylhexyl)-phthalate from commonly used intravenous infusion tubing and bags. Adverse reactions to the plasticizers have been reported, such as respiratory distress, necessitating the use of special infusion systems at extra expense and time. Waugh, et al. (1991) Am J Hosp Pharmacists 48:1520.
In light of these problems, it can be seen that the ideal emulsion vehicle would be inexpensive, non-irritating or even nutritive and palliative in itself, terminally sterilizable by either heat or filtration, stable for at least 1 year under controlled storage conditions, accommodate a wide variety of water insoluble and poorly soluble drugs and be substantially ethanol-free. In addition to those drugs which are lipophilic and dissolve in oils, also needed is a vehicle which will stabilize, and carry in the form of an emulsion, drugs which are poorly soluble in lipids and in water.